Blunt hepatic injury: a paradigm shift from operative to nonoperative management in the 1990s.

OBJECTIVE: To analyze the outcome of hemodynamically stable patients with blunt hepatic injury managed nonoperatively, and to examine the impact of this approach on the outcome of all patients with blunt hepatic injury. SUMMARY BACKGROUND DATA: Until recently, operative management has been the standard for liver injury. A prospective trial from the authors' institution had shown that nonoperative management could safely be applied to hemodynamically stable patients with blunt hepatic injury. The present study reviewed the authors' institutional experience with blunt hepatic trauma since that trial and compared the results with prior institutional experience. METHODS: Six hundred sixty-one patients with blunt hepatic trauma during the 5-year period ending December 1998 were reviewed (NONOP2). The outcomes were compared with two previous studies from this institution: operative 1985 to 1990 (OP) and nonoperative 1993 to 1994 (NONOP1). RESULTS: All 168 OP patients were managed operatively. Twenty-four (18%) of 136 NONOP1 patients and 101 (15%) of the 661 NONOP2 patients required immediate exploration for hemodynamic instability. Forty-two (7%) patients failed nonoperative management; 20 were liver-related. Liver-related failures of nonoperative management were associated with higher-grade injuries and with larger amounts of hemoperitoneum on computed tomography scanning. Twenty-four-hour transfusions, abdominal infections, and hospital length of stay were all significantly lower in the NONOP1 and NONOP2 groups versus the OP cohort. The liver-related death rate was constant at 4% in the three cohorts over the three time periods. CONCLUSIONS: Although urgent surgery continues to be the standard for hemodynamically compromised patients with blunt hepatic trauma, there has been a paradigm shift in the management of hemodynamically stable patients. Approximately 85% of all patients with blunt hepatic trauma are stable. In this group, nonoperative management significantly improves outcomes over operative management in terms of decreased abdominal infections, decreased transfusions, and decreased lengths of hospital stay.

Early and late acute respiratory distress syndrome: two distinct clinical entities.

BACKGROUND: Despite numerous advances in surgical critical care and ventilatory management, mortality rates for acute respiratory distress syndrome (ARDS) have remained relatively constant. Pressure-limited and non-pressure-limited ventilatory techniques have been advocated with disparate results. We hypothesized that there are two forms of ARDS, which may account for the conflicting clinical reports. METHODS: Patients with posttraumatic ARDS were identified and reviewed. ARDS was defined as PaO2/FiO2 ratio less than 200 with diffuse bilateral infiltrates on chest radiograph and no congestive heart failure. Patients were analyzed relative to injury mechanism, transfusions, fluid balance, presence of pneumonia (defined as > or =10(5) colony-forming units/mL in bronchoalveolar lavage effluent), and outcome. All were managed with a non-pressure-limited strategy. RESULTS: During a 5.5-year period, 178 patients with posttraumatic ARDS were identified. Mean Injury Severity Score and age were 29 and 40 years, respectively. Patients were stratified by time of ARDS diagnosis. Eighty-two patients (46%) had early ARDS (within 48 hours after admission), and 96 patients (54%) had late ARDS (>48 hours between admission and diagnosis). There were no differences in Injury Severity Score, but the late group was significantly older. The early ARDS group was characterized by profound hemorrhagic shock and had significant differences from the late group in incidence of penetrating injury (30 vs. 10%; p<0.001), admission base deficit (-7.7 vs. -4.2 mEq/L; p<0.001), 48-hour transfusions (19.7 vs. 9.4; p<0.0001), initial 5-day fluid balance (19.9 vs. 10.1 L; p<0.0001), and initial PaO2/FiO2 (121 vs. 141; p<0.007). Pneumonia before ARDS was significantly associated with late ARDS (38 vs. 9%; p<0.001). ARDS-related mortality was primarily caused by hemorrhagic shock in the early group and progressive multiple organ failure in the late group. CONCLUSION: There are two distinct forms of posttraumatic ARDS. Early ARDS is characterized by hemorrhagic shock with capillary leak. Late ARDS frequently follows pneumonia and is associated with multiple system injury. Further studies should differentiate between these two distinct syndromes.

Treatment of acute anticholinergic poisoning with physostigmine.

Five cases of acute anticholinergic poisoning presenting to an inner-city emergency department (ED) are discussed. All five patients presented with classic signs and symptoms of anticholinergic toxicity, which included tachycardia, hot, dry and flushed skin, markedly dilated and fixed pupils, and pronounced delirium. The patients were violently agitated, and physical restraint was required. Initial treatment with benzodiazepines did not diminish their combative behavior. Treatment with intravenous physostigmine salicylate resulted in a decrease in agitation within 15 to 20 minutes of therapy. No untoward effects occurred as a result of treatment with physostigmine.

Preemptive pain control in patients having laparoscopic hernia repair: a comparison of ketorolac and ibuprofen.

OBJECTIVES: To determine if nonsteroidal anti-inflammatory drugs provide adequate pain control for patients having laparoscopic hernia repair and to compare the effectiveness of ketorolac tromethamine with ibuprofen in reducing postoperative laparoscopic hernia pain. DESIGN AND SETTING: Prospective double-blind randomized study at a 100-bed community hospital. PATIENTS: Seventy patients ranging in age from 16 to 83 years scheduled for elective laparoscopic inguinal hernia repair. INTERVENTIONS: Patients undergoing laparoscopic hernia repair were enrolled in a double-blind randomized study to compare the 2 treatments. Group 1 received a placebo capsule 1 hour before surgery and ketorolac tromethamine, 60 mg intravenously, at the time of trocar insertion. Group 2 received ibuprofen, 800 mg an hour before surgery, and isotonic sodium chloride solution, 2 mL intravenously, at the time of trocar insertion. In addition, all patients received local infiltration of 30 mL of bupivacaine hydrochloride into their trocar sites. All patients were discharged within 5 hours of the operation and were instructed to take 400 mg of ibuprofen orally every 4 hours for 24 hours whether or not they were experiencing pain. A 24-hour supply of ibuprofen was provided to all study patients. Pain was assessed using the Visual Analog Pain Scale with a maximum pain rating of 100. Assessments were done at the time of and 18 hours after discharge. MAIN OUTCOME MEASURE: Postoperative pain 18 and 24 hours after discharge was assessed using a standardized questionnaire in a telephone interview by a registered nurse from the Outpatient Surgical Unit. RESULTS: There was no significant difference in the level of pain experienced by 35 patients who received ketorolac intravenously and 35 who received ibuprofen orally. There was no significant difference between the 2 treatment groups in the amount of pain experienced at discharge and 18 hours after discharge. CONCLUSIONS: Pain relief from ibuprofen, 800 mg, administered orally an hour before laparoscopic hernia repair was not statistically different from that obtained with intravenous ketorolac, 60 mg, administered intraoperatively when comparing the hospital discharge pain score and the mean and highest pain scores 18 hours after discharge. Ibuprofen offers equivalent pain control at a lower cost and reduced potential for adverse drug events compared with intravenous ketorolac in patients having laparoscopic hernia repair. No patient required narcotic supplementation, and pain control was judged satisfactory by all the patients.

Splanchnic and systemic hemodynamic responses to portal vein endotoxin after resuscitation from hemorrhagic shock.

BACKGROUND: Hemorrhagic shock and sepsis are usually studied separately or in rodents. This study combined the two insults in a large animal model. METHODS: Anesthetized pigs were bled, held in shock for 1 hour, and then resuscitated with fluid. After 3 days, Escherichia coli endotoxin LPS was infused into the portal vein (150 micrograms/kg x 30 min) to mimic the effect of enteric substances breaching the mucosal barrier. Systemic and splanchnic hemodynamics, circulating leukocytes, and plasma levels of tumor necrosis factor (TNF) were measured in five groups: 40% hemorrhage plus fluid only resuscitation, 40% hemorrhage plus fluid-blood resuscitation, 50% hemorrhage plus fluid-blood resuscitation, sham, or sham plus 5 micrograms/kg LPS priming dose instead of hemorrhage. RESULTS: On day 4 before infusion of LPS, there were no differences between groups in hemodynamics or O2 utilization, but systemic O2 delivery and O2 consumption were each reduced in the hemorrhaged groups because of the hemodilution associated with resuscitation. For 3 hours after infusion of LPS, all animals received aggressive fluid and respiratory support, but arterial blood pressure decreased, and systemic O2 utilization, splanchnic O2 utilization, and arterial lactate level increased; there were no differences between groups. In the 50% group compared with sham, LPS-evoked decreases in cardiac index and stroke index were eliminated, and LPS-evoked tachycardia, pulmonary and systemic vasoconstriction, and increases in hepatic and portal vein lactate levels were blunted. Despite similar leukocyte counts before infusion of LPS and similar leukopenia after LPS infusion, plasma LPS level was higher in the 50% group compared with sham. Furthermore, LPS evoked increases in portal and hepatic vein plasma TNF in the shams, but those changes were reduced in the 50% group. CONCLUSIONS: Most responses to LPS were similar after hemorrhagic shock or a sham operation, which is inconsistent with the concept of "priming." LPS-evoked increases in plasma TNF were blunted after shock, probably because of trauma-induced immune dysfunction. A combined shock plus septic challenge in a large animal model may be valuable for investigating the pathogenic mechanism in human beings.

The association of cigarette smoking with clinical subtypes of ophthalmic Graves' disease.

We retrospectively reviewed the smoking habits of 124 patients with Graves' ophthalmopathy. We divided these patients into those without restrictive myopathy (type I) and those with restrictive myopathy (type II). Sixty-three percent of type I and 83% of type II patients were smokers at the time of diagnosis. The incidence of smoking in an age- and gender-matched randomly selected control population in the same geographic area was 30% for controls matched to our type I group and 26.6% for controls matched to our type II group. We believe smoking is an important factor in the development of both clinical subtypes of Graves' ophthalmopathy.

Source of missing doses in a decentralized unit dose system: a quality assurance review.

Unit dose drug distribution systems have become the prominent drug delivery system in healthcare institutions. Much of the success of the unit-dose concept can be attributed to the increased accountability of medications dispensed and administered. However, as a direct result of this added accountability the problem of "missing doses" has become more evident than under non-unit dose systems. A quality assurance audit was conducted to identify the reasons for missing doses and identify the responsibility for them. A missing dose record was developed through a collaborative effort of the pharmacy and nursing departments. Both staff nurses and pharmacists were invited to review and suggest modifications prior to using the form. All potential reasons for missing dose occurrences were listed on the form. When a missing dose occurred the pharmacist and the nurse completed form jointly to ensure agreement as to the cause. During the 25 day study period the pharmacy dispensed 54,082 doses of medications to patients on the study floors. During that period 227 incidents of missing doses were documented. This resulted in an overall incidence of missing doses of 0.4%. Of the 227 missing doses, a reason for the occurrence was identified for 170 (75%). Nine classes of medications accounted for 62% of all reported missing doses. When responsibility for the missing doses was examined 13.3% were pharmacy generated and 45.8% were nursing generated. The results of the audit generated several suggestions which may decrease the number of missing doses and further improve our drug distribution service.